In this paper, we propose a novel algorithm to optimize the energy-efficiency(EE) of orthogonal frequency division multiplexing-based cognitive radiosystems under channel uncertainties. We formulate an optimization problem thatguarantees a minimum required rate and a specified power budget for thesecondary user (SU), while restricting the interference to primary users (PUs)in a statistical manner. The optimization problem is non-convex and it istransformed to an equivalent problem using the concept of fractionalprogramming. Unlike all related works in the literature, we consider the effectof imperfect channel-stateinformation (CSI) on the links between the SUtransmitter and receiver pairs and we additionally consider the effect oflimited sensing capabilities of the SU. Since the interference constraints aremet statistically, the SU transmitter does not require perfect CSI feedbackfrom the PUs receivers. Simulation results sho w that the EE deteriorates asthe channel estimation error increases. Comparisons with relevant works fromthe literature show that the interference thresholds at the PUs receivers canbe severely exceeded and the EE is slightly deteriorated if the SU does no taccount for spectrum sensing errors.
展开▼
